--- trunk/libf/phylmd/Mobidic/regr_pr_coefoz.f90 2008/08/05 13:31:32 17 +++ trunk/libf/phylmd/Mobidic/regr_pr_coefoz.f90 2012/04/20 14:58:43 61 @@ -1,5 +1,14 @@ module regr_pr_coefoz + ! Both procedures of this module read a single Mobidic ozone + ! coefficient from "coefoz_LMDZ.nc", at the current day, regrid this + ! coefficient in pressure to the LMDZ vertical grid and pack it to the LMDZ + ! horizontal "physics" grid. + ! The input data is a 2D latitude -- pressure field. + ! The target horizontal LMDZ grid is the "scalar" grid: "rlonv", "rlatu". + ! We assume that the input data is already on the LMDZ "rlatu" + ! latitude grid. + implicit none contains @@ -8,20 +17,21 @@ ! "regr_pr_av_coefoz" stands for "regrid pressure averaging ! coefficient ozone". - ! This procedure reads a single Mobidic ozone coefficient from - ! "coefoz_LMDZ.nc", at the current day, regrids this parameter in - ! pressure to the LMDZ vertical grid and packs it to the LMDZ - ! horizontal "physics" grid. - ! Regridding in pressure is done by averaging a step function. + ! The target vertical LMDZ grid is the grid of layer boundaries. + ! The input data does not depend on longitude, but the pressure + ! at LMDZ layers does. + ! Therefore, the values on the LMDZ grid do depend on longitude. + ! Regridding in pressure is done by averaging a step function of pressure. use dimens_m, only: iim, jjm, llm use dimphy, only: klon - use netcdf95, only: nf95_inq_varid, handle_err - use netcdf, only: nf90_get_var use grid_change, only: dyn_phy - use regr_pr, only: regr_pr_av - use numer_rec, only: assert + use netcdf, only: nf90_get_var + use netcdf95, only: nf95_inq_varid, handle_err + use nr_util, only: assert + use numer_rec_95, only: regr1_step_av use press_coefoz_m, only: press_in_edg + use pressure_var, only: p3d integer, intent(in):: ncid ! NetCDF ID of the file character(len=*), intent(in):: name ! of the NetCDF variable @@ -30,11 +40,11 @@ real, intent(out):: v3(:, :) ! (klon, llm) ! (ozone coefficient from Mobidic on the "physics" grid) ! ("v3(i, k)" is at longitude "xlon(i)", latitude - ! "xlat(i)", middle of layer "k".) + ! "xlat(i)", in layer "k".) ! Variables local to the procedure: - integer varid, ncerr - integer k + + integer varid, ncerr ! for NetCDF real v1(jjm + 1, size(press_in_edg) - 1) ! (ozone coefficient from "coefoz_LMDZ.nc" at day "julien") @@ -42,9 +52,11 @@ ! pressure interval "[press_in_edg(k), press_in_edg(k+1)]".) real v2(iim + 1, jjm + 1, llm) - ! (ozone parameter from Mobidic on the "dynamics" grid) - ! "v2(i, j, k)" is at longitude "rlonv(i)", latitude - ! "rlatu(j)", middle of layer "k".) + ! (ozone coefficient on the "dynamics" grid) + ! ("v2(i, j, k)" is at longitude "rlonv(i)", latitude + ! "rlatu(j)" and for pressure interval "[p3d(i, j, k+1), p3d(i, j, k)]".) + + integer i, j, k !-------------------------------------------- @@ -55,12 +67,21 @@ ! Get data at the right day from the input file: ncerr = nf90_get_var(ncid, varid, v1, start=(/1, 1, julien/)) call handle_err("regr_pr_av_coefoz nf90_get_var " // name, ncerr, ncid) - ! Latitudes are in increasing order in the input file while - ! "rlatu" is in decreasing order so we need to invert order: + ! Latitudes are in ascending order in the input file while + ! "rlatu" is in descending order so we need to invert order: v1 = v1(jjm+1:1:-1, :) ! Regrid in pressure at each horizontal position: - v2 = regr_pr_av(v1, press_in_edg) + do j = 1, jjm + 1 + do i = 1, iim + if (dyn_phy(i, j)) then + v2(i, j, llm:1:-1) & + = regr1_step_av(v1(j, :), press_in_edg, & + p3d(i, j, llm+1:1:-1)) + ! (invert order of indices because "p3d" is in descending order) + end if + end do + end do forall (k = 1:llm) v3(:, k) = pack(v2(:, :, k), dyn_phy) @@ -70,20 +91,23 @@ subroutine regr_pr_int_coefoz(ncid, name, julien, top_value, v3) - ! This procedure reads a single Mobidic ozone coefficient from - !"coefoz_LMDZ.nc", at the current day, regrids this parameter in - ! pressure to the LMDZ vertical grid and packs it to the LMDZ - ! horizontal "physics" grid. + ! "regr_pr_int_coefoz" stands for "regrid pressure interpolation + ! coefficient ozone". + ! The target vertical LMDZ grid is the grid of mid-layers. + ! The input data does not depend on longitude, but the pressure + ! at LMDZ mid-layers does. + ! Therefore, the values on the LMDZ grid do depend on longitude. ! Regridding is by linear interpolation. use dimens_m, only: iim, jjm, llm use dimphy, only: klon - use netcdf95, only: nf95_inq_varid, handle_err - use netcdf, only: nf90_get_var use grid_change, only: dyn_phy - use regr_pr, only: regr_pr_int - use numer_rec, only: assert + use netcdf, only: nf90_get_var + use netcdf95, only: nf95_inq_varid, handle_err + use nr_util, only: assert + use numer_rec_95, only: regr1_lint use press_coefoz_m, only: plev + use pressure_var, only: pls integer, intent(in):: ncid ! NetCDF ID of the file character(len=*), intent(in):: name ! of the NetCDF variable @@ -93,22 +117,24 @@ ! (extra value of ozone coefficient at 0 pressure) real, intent(out):: v3(:, :) ! (klon, llm) - ! (ozone parameter from Mobidic on the "physics" grid) + ! (ozone coefficient from Mobidic on the "physics" grid) ! ("v3(i, k)" is at longitude "xlon(i)", latitude ! "xlat(i)", middle of layer "k".) ! Variables local to the procedure: - integer varid, ncerr - integer k + + integer varid, ncerr ! for NetCDF real v1(jjm + 1, 0:size(plev)) ! (ozone coefficient from "coefoz_LMDZ.nc" at day "julien") ! ("v1(j, k >=1)" is at latitude "rlatu(j)" and pressure "plev(k)".) real v2(iim + 1, jjm + 1, llm) - ! (ozone parameter from Mobidic on the "dynamics" grid) + ! (ozone coefficient on the "dynamics" grid) ! "v2(i, j, k)" is at longitude "rlonv(i)", latitude - ! "rlatu(j)", middle of layer "k".) + ! "rlatu(j)" and pressure "pls(i, j, k)".) + + integer i, j, k !-------------------------------------------- @@ -119,15 +145,23 @@ ! Get data at the right day from the input file: ncerr = nf90_get_var(ncid, varid, v1(:, 1:), start=(/1, 1, julien/)) call handle_err("regr_pr_int_coefoz nf90_get_var " // name, ncerr, ncid) - ! Latitudes are in increasing order in the input file while - ! "rlatu" is in decreasing order so we need to invert order: + ! Latitudes are in ascending order in the input file while + ! "rlatu" is in descending order so we need to invert order: v1(:, 1:) = v1(jjm+1:1:-1, 1:) ! Complete "v1" with the value at 0 pressure: v1(:, 0) = top_value ! Regrid in pressure at each horizontal position: - v2 = regr_pr_int(v1, (/0., plev/)) + do j = 1, jjm + 1 + do i = 1, iim + if (dyn_phy(i, j)) then + v2(i, j, llm:1:-1) & + = regr1_lint(v1(j, :), (/0., plev/), pls(i, j, llm:1:-1)) + ! (invert order of indices because "pls" is in descending order) + end if + end do + end do forall (k = 1:llm) v3(:, k) = pack(v2(:, :, k), dyn_phy)